Abstract: To address the need for in-situ characterization of extraterrestrial regolith in deep space exploration, this study quantifies the correlation between high-frequency passive features in signals and particle size under low-velocity impact scenarios on planetary surfaces, such as landing, penetration, and sampling. The feasibility of utilizing these features for particle-scale identification is explored. A spherical impactor equipped with built-in accelerometers was designed, and impact experiments were conducted on Earth using dry sand with varying particle sizes and relative densities, providing calibration and validation data for the identification method. Time–frequency spectral analysis was performed on the acceleration signals. The rigid-body acceleration signal a_rigid(t) was used to normalize the original acceleration signal, yielding a dimensionless fluctuation signal a_fluc(t). Based on the power spectral density of a_fluc(t), a fluctuation energy metric E was obtained, and a fluctuation energy coefficient χ was defined to quantify the degree of acceleration signal fluctuation. Using the experimental data, an empirical relationship between the ratio of the spherical impactor diameter D to the mean particle diameter \bard and the fluctuation energy coefficient χ was established, and supplementary validation experiments were performed. The results show that the particle size d_g is the dominant granular medium parameter influencing the dimensionless fluctuation signal a_fluc(t); the relative density Dr mainly affects the rigid-body acceleration a_rigid(t) and its peak value, but has little influence on a_fluc(t); and when the diameter ratio between the impactor and the mean particle size is less than or equal to 30, the fluctuation energy coefficient χ exhibits good sensitivity for identifying the mean particle size \bard . The above results indicate that the fluctuation characteristics in the acceleration signal and their calibration relationships can serve as auxiliary diagnostic indicators for identifying the particle scale of extraterrestrial regolith, supporting the interpretation of acceleration data obtained during contact processes.